This invention relates to the compositions of matter useful as catalysts, to a method for preparing these catalysts and to a method for polymerization utilizing the catalysts.
The use of soluble Ziegler-Natta type catalysts in the polymerization of olefins is well known in the prior art. In general, such systems include a Group IV-B metal compound and a metal or metalloid alkyl cocatalyst, such as aluminum alkyl cocatalyst. More broadly, it may be said to include a mixture of a Group I-III metal alkyl and a transition metal complex from Group IVB-VB metals, particularly titanium, zirconium, or hafnium with aluminum alkyl cocatalysts.
First generation cocatalyst systems for homogeneous metallocene Ziegler-Natta olefin polymerization, alkylaluminum chlorides (AlR.sub.2 Cl), exhibit low ethylene polymerization activity levels and no propylene polymerization activity. Second generation cocatalyst systems, utilizing methyl aluminoxane (MAO), raise activities by several orders of magnitude. In practice however, a large stoichiometric excess of MAO over catalyst ranging from several hundred to ten thousand must be employed to have good activities and stereoselectivities. Moreover, it has not been possible to isolate characterizable metallocene active species using MAO. The third generation of cocatalyst, B(C.sub.6 F.sub.5).sub.3, proves to be far more efficient while utilizing a 1:1 catalyst-cocatalyst ratio. Although active catalyst species generated with B(C.sub.6 F.sub.5).sub.3 are isolable and characterizable, the anion MeB(C.sub.6 F.sub.5).sub.3.sup.- formed after Me.sup.- abstraction from metallocene dimethyl complexes is weakly coordinated to the electron-deficient metal center, thus resulting in a drop of certain catalytic activities. The recently developed B(C.sub.6 F.sub.5).sub.4.sup.- type of non-coordinating anion exhibits some of the highest reported catalytic activities, but such catalysts have proven difficult to obtain in the pure state due to poor thermal stability and poor crystallizability, which is crucial for long-lived catalysts and for understanding the role of true catalytic species in the catalysis for the future catalyst design. Synthetically, it also takes two more steps to prepare such an anion than for the neutral organo-Lewis acid.